Open Steffy-zxf opened 1 year ago
Follow the tutorial to deploy DB-GPT on the ubuntu machine, 24g single-card GPU. As I continued to ask questions, CUDA memory kept increasing until memory exploded. How to free video memory?
During inference, the video memory itself will grow,We are also looking for a solution.
Hi, any solution for this issue? I have the same issue and need to always restart llm server
Follow the tutorial to deploy DB-GPT on the ubuntu machine, 24g single-card GPU. As I continued to ask questions, CUDA memory kept increasing until memory exploded. How to free video memory?